Patients with evidence of spontaneous anti-tumor T cell responses have a better prognosis and are likely to respond to checkpoint blockade immunotherapies currently used in the clinic. Identifying why some patients lack spontaneous anti-tumor immune responses and finding ways to promote endogenous responses will likely expand efficacy to these patients. Our lab has previously shown immunogenic tumors spontaneously activate the innate immune system through the STING pathway. The STING pathway senses cytosolic DNA, which activates a signal transduction pathway culminating in nuclear translocation of transcription factors IRF3 and NF-?B that, in turn, induce expression of several genes including IFN-?. STING signaling and IFN-? receptor signaling in tumor-infiltrating immune cells are both required for optimal priming of CD8+ T cells against tumor antigens. Likewise, intratumoral injection of STING agonists increases IFN-? expression, anti-tumor T cell priming, and causes dramatic tumor rejection in pre-clinical models. As part of our work with intratumoral STING agonists we observed that the tumor cells themselves were the only cell type present in the tumor microenvironment unable to express IFN-? in response to cytosolic DNA or direct STING agonists. We tested a range of tumor cell lines and found the vast majority was unable to express IFN-? downstream of STING signaling, arguing that loss of activation of this pathway might occur regularly as a component of oncogenesis. Our over-arching hypothesis is that determining why tumor cells lack IFN-? expression and reversing this phenotype could lead to a new strategy to induce endogenous anti-tumor immune responses. Our preliminary data indicates that STING signaling is largely in tact in tumor cells up to and including nuclear translocation of IRF3. We find tumor cells have a defect in IRF3 DNA binding at the IFN-? locus and that the lack of IFN-? expression following STING activation is a dominant phenotype. To better understand tumor cell-intrinsic STING signaling, and how it is defective in tumor cells, we propose to: 1) identify and reverse the molecular mechanism preventing STING induction of IFN-? in tumor cells using pre-clinical models where we can observe how this affects the tumor/immune interface in vivo 2) understand the potential for STING signaling in human cancer by analyzing both cell lines and tumor biopsies. Preliminary data suggests tumor cells differ from control cells in the epigenetic accessibility of the IFN locus and NF-?B signaling after STING activation. Therefore we will study NF-kB signaling in tumor cells and it's affect on accessibility of the IFN locus. We will also silence the STING pathway in the few tumor cells capable of expressing IFN-? to observe how this modulates the interface with host immunity. As we analyze human tumor cell lines and biopsies we will begin to determine what fraction of patients might respond to therapies designed to enable tumor cells to express IFN-?. Ultimately our work will be an important part of a growing body of evidence indicating a deeper connection between oncogenesis and innate immunity that implicates the latter as a tumor suppressor. !